Position estimation in indoor using networked GNSS sensors and a range-azimuth sensor

Abstract

The global navigation satellite systems (GNSS) receivers suffer from determining an accurate position estimate in the indoor region due to non-line of sight (Non-LOS) conditions. This paper proposes a novel method of position estimation within the indoor region by using distributed GNSS sensors and a range-azimuth sensor setup. All the GNSS sensors are connected to a central node; the inaccurate position estimates evolved from the Kalman filter (KF) framework are transmitted to a central node as primary data. The deviation between the inaccurate position estimate and the GNSS sensor’s actual position is the positional deviation (PD) vector, which needs to be estimated. In the same indoor region, a range-azimuth sensor is also deployed. It estimates the GNSS sensor’s physical location in its local coordinates, and these estimates are being transmitted to a central node as secondary data. Further, by using the primary and secondary data, we formulated a PD vector compensation followed by a sequential fusion (SF) framework to derive the precise locations of both GNSS sensors and the range-azimuth sensor by recursively estimating the PD vector. Finally, the Cramer–Rao lower bound (CRLB) for the proposed framework is derived. The simulation results are quantified with the root mean square error (RMSE) and CRLB.